A new NBER paper on the Wuhan lockdown

We quantify the causal impact of human mobility restrictions, particularly the lockdown of the city of Wuhan on January 23, 2020, on the containment and delay of the spread of the Novel Coronavirus (2019-nCoV). We employ a set of difference-in-differences (DID) estimations to disentangle the lockdown effect on human mobility reductions from other confounding effects including panic effect, virus effect, and the Spring Festival effect. We find that the lockdown of Wuhan reduced inflow into Wuhan by 76.64%, outflows from Wuhan by 56.35%, and within-Wuhan movements by 54.15%. We also estimate the dynamic effects of up to 22 lagged population inflows from Wuhan and other Hubei cities, the epicenter of the 2019-nCoV outbreak, on the destination cities’ new infection cases. We find, using simulations with these estimates, that the lockdown of the city of Wuhan on January 23, 2020 contributed significantly to reducing the total infection cases outside of Wuhan, even with the social distancing measures later imposed by other cities. We find that the COVID-19 cases would be 64.81% higher in the 347 Chinese cities outside Hubei province, and 52.64% higher in the 16 non-Wuhan cities inside Hubei, in the counterfactual world in which the city of Wuhan were not locked down from January 23, 2020. We also find that there were substantial undocumented infection cases in the early days of the 2019-nCoV outbreak in Wuhan and other cities of Hubei province, but over time, the gap between the officially reported cases and our estimated “actual” cases narrows significantly. We also find evidence that enhanced social distancing policies in the 63 Chinese cities outside Hubei province are effective in reducing the impact of population inflows from the epicenter cities in Hubei province on the spread of 2019-nCoV virus in the destination cities elsewhere.

That is by Hanming Fang, Long Wang, and Yang Yang, of course do beware data quality issues.


Data quality is just another word for mood affiliation, and data don't mean nothin' hon' if it ain't free, no no.

This paper is probably not very useful because it overlooks the fact that in addition to lockdowns, China mass-tested millions and isolated lots of cases. https://www.courthousenews.com/china-goes-door-to-door-in-wuhan-seeking-infections/

Good point, and that's why the Trump Jan. 31, 2020 ban on all non-US citizens flying into the USA from China would not have worked. By that date, already infected had already entered the USA (and you need to test to find them): 2/1/20 : (WaPo on the Trump bar on non-US citizens from CHina): "Shortly after the White House announced the new restrictions and said there were six confirmed U.S. cases, a seventh case was confirmed in Santa Clara County, Calif."

Bonus trivia: 'the infected'...I saw for the first time the other day the excellent virus Zombie movie "28 Days Later". It had some plot twists I was not expecting (I thought for sure one of the protagonists would sacrifice their life for the other), and the empty street scenes were great. It was not as scary as online reviews said (I don't like scary movies and fake blood, though my hot Filipina girl does, lol)

“ that's why the Trump Jan. 31, 2020 ban on all non-US citizens flying into the USA from China would not have worked.”

At least he was ahead of the curve, and tried something. Can you imagine the relentless hysteria over Trump if he’d done nothing?

At least we were spared the relentless hysteria over Trump, as if he’s personally responsible for FDA and CDC decisions and policy, and as if he’s got “blood on his hands.” Whew!

>Can you imagine the relentless hysteria...

There is really no need to "imagine" relentless hysteria.

Yet he was still behind Italy. As is the U.S. at this point in terms of the coronavirus pandemic, but that is going to change, unfortunately.

Since March 23 when Ray Lopez touted SPDN (a 1x Inverse S&P ETF and frankly a pretty weak choice if you want to risk getting your face ripped off in this market), the stock is down 19.5%.
If you followed his advice you are down as much in a week as the S&P is for the month of March.
So Ray Lopez is officially a contrarian indicator for the US stock market

+1. Nice handle

I bought into the market when Bill Gates said countries that shut down for coronavirus could bounce back in weeks. It appears his observation on March 18 subsequently turned the stock market around.

When did economists become epidemiologists?

Same time “bad data” became a way of not saying “CPC lies.”

Things I think I can infer by skimming this paper:

1. China's success in cities outside of Hubei worked primarily through the restriction of outflow of Hubei residents, secondarily through social distancing to limit local-to-local transmission.

2. For other counties, travel restrictions for people who had been in Hubei should have begun earlier. (In the US, the restrictions began February 1, well after the first confirmed case in Everett, WA on January 20.)

3. For other countries, social distancing should have begun earlier, perhaps as early as the first confirmed case in the region.

What reliability can we attach to Chinese data at this point?

Suppose the Party lies 20% of the time (underreporting etc). For those times when projections really matter, does the Party have a second set of “truer” data? In other words, surely the Politburo knows roughly how things will really play out?

Not necessarily. This is a good question. I've wondered if China has become more reality based since the days of the Great Leap Forward, which killed 50 million Chinese because Mao actually believed Lysenko's fake science about grain yields. They actually sold grain reserves to the USSR in the midst of famine. I assume things have gotten better since those days but wonder if the sloth by provincial party members in Wuhan to sound an alarm was a relic of those days, i.e. not wanting to tell party higher ups bad news.

I also wonder if, when the whole tale of the CDC/FDA testing fiasco is told, that a similar dynamic with Trump (that the underlings were timid in shooting up flares because they were afraid to draw flak) will also be part of the story. Not to say that Tyler's preferred narrative of CDC/FDA protecting bureaucratic prerogatives, being overly cautious about test precision etc isn't also true too.

I seem to remember something going on in January that consumed Washington. I wonder what that was? Hmmm.

Impeachments have consequences.

It's been well known for years that local, provincial, *and* national level officials in China will both inflate and smooth economic numbers to get to the growth numbers that they want. A quick Google search finds plenty of articles on that topic.

I couldn't find a specific link to the quote, but there was something much cited 5-10 years ago of a senior Chinese official telling a Western official that he looked at certain metrics like electricity consumption, because he knew that provincial leaders were manipulating economic growth numbers to hit targets and enhance their personal chances for promotion. Wish I could provide a more specific confirmation, but my recollection is that it was pretty well-sourced, with a named Western official being quoted about a named Chinese official. (In other words, it wasn't just "sources familiar with the matter" anonymous sourcing.)

The paper is more about the people in and out of Wuhan but I note that they also did their quarantine a little differently
--Early February they quarantined the mild cases (those that don't require hospitalization) in makeshift hospitals converted from offices, stadiums and gymnasiums. They removed them from the family home. They realized otherwise all family members were getting infected. Italy and the US are not doing this I believe.
--They made masks mandatory anywhere outside the home, no exceptions and people even wore them indoors

Are we supposed to take info such as:

"over time, the gap between the officially reported cases and our estimated “actual” cases narrows significantly"

seriously? I suppose it could be true in % sense, after all, 50 cases with 0 reported the gap would be infinite; 800k cases with 80k reported would only be a gap of 90%.

Hilarious names on all three of the study's authors. Especially when said in succession.

You wish your name was Long Wang.

Comments for this post are closed